Effect of High-Pressure−Moderate-Temperature Processing on the

9184

J. Agric. Food Chem. 2006, 54, 9184−9192

Effect of High-Pressure−Moderate-Temperature Processing on the Volatile Profile of Milk PEDRO A. VAZQUEZ-LANDAVERDE, J. ANTONIO TORRES,

AND

MICHAEL C. QIAN*

Department of Food Science and Technology, Oregon State University, Corvallis, Oregon 97331

The effects of high hydrostatic pressure on volatile generation in milk were investigated in this study. Raw milk samples were treated under different pressures (482, 586, and 620 MPa), temperatures (25 and 60 °C), and holding times (1, 3, and 5 min). Samples submitted to heat treatments alone (25, 60, and 80 °C for 1, 3, and 5 min) were used for comparison. Trace volatile sulfur compounds were analyzed using solid-phase microextraction (SPME) and gas chromatography (GC) with pulsedflame photometric detection (PFPD), whereas the rest of the volatile compounds were analyzed using SPME-GC with flame ionization detection (FID). Multivariate analysis of variance (MANOVA) and principal component analysis (PCA) were used to study the effect of pressure, temperature, and time on volatile generation. Relative concentration increases of 27 selected volatile compounds were compared to an untreated sample. It was found that pressure, temperature, and time, as well as their interactions, all had significant effects (P < 0.001) on volatile generation in milk. Pressure and time effects were significant at 60 °C, whereas their effects were almost negligible at 25 °C. The PCA plot indicated that the volatile generation of pressure-heated samples at 60 °C was different from that of heated-alone samples. Heat treatment tended to promote the formation of methanethiol, hydrogen sulfide, methyl ketones, and aldehydes, whereas high-pressure treatment favored the formation of hydrogen sulfide and aldehydes. KEYWORDS: Milk; high-pressure processing; volatile; volatile sulfur compounds

INTRODUCTION

Thermal processing is the prevailing method to achieve microbial safety and shelf-life stability of milk. Although hightemperature-short-time (HTST) pasteurization (typically at 72 °C for 15 s) is typically used commercially to process milk, the product shelf life is only 20 days at refrigeration temperatures. Ultrahigh-temperature (UHT) processing (135-150 °C for 3-5 s) produces a product that is stable at room temperature for up to 6 months; however, this process can induce strong “cooked” off-aroma notes in milk (1), thus limiting its marketing in the United States and many other countries (2). Numerous studies have identified volatile sulfur compounds, aldehydes, and methyl ketones as the most important contributors to this “cooked” off-aroma defect (3-9), and reliable quantification methods have been developed to analyze these off-aroma compounds (8-11). New technologies are needed to process milk without compromising its flavor. Several nonthermal processing technologies have been explored to achieve microbial safety and minimize off-flavor formation. Microfiltration using cross-flow membrane separation has showed promising results in eliminat* Address correspondence to this author at 100 Wiegand Hall, Department of Food Science and Technology, Oregon State University, Corvallis, OR 97331 [telephone (541) 737-9114; fax (541) 737-1877; e-mail [email protected]].

ing bacteria from milk and increasing shelf life without the development of off-flavors (12, 13). However, high levels of milk fat could foul the membrane and place some restrictions on the use of microfiltration as an alternative technique for milk processing. High hydrostatic pressure processing (HPP), a new technology to the food industry (14), can destroy microorganisms by high hydrostatic pressure without heat (15-17). This technology has been gaining commercial acceptance in the manufacture of food products with “fresh” flavor that are not possible with other preservation technologies (18, 19). To retain the “fresh” milk flavor, HPP has been studied as a potential alternative for the pasteurization of milk. A microbiological reduction similar to that of pasteurized milk has been achieved using pressure treatments of 400 MPa for 15 min or 500 MPa for 3 min at room temperature (20). At moderate temperature (55 °C), HPP (586 MPa for 5 min) can significantly extend the shelf life of milk beyond 45 days, which far exceeds that of pasteurized milk (21). HPP has been reported to change some properties of the foods. HPP can reduce the size of casein micelles in milk at pressures above 230 MPa, resulting in a decrease in whiteness and turbidity and an increase in the viscosity of milk (22). High pressure can also affect the crystallization properties of milk fat. The crystallization behavior of milk fat can be altered because the high pressure will shift the phase transition

10.1021/jf061497k CCC: $33.50 © 2006 American Chemical Society Published on Web 10/28/2006

J. Agric. Food Chem., Vol. 54, No. 24, 2006

Volatile Profile of Milk temperature (23). It is generally assumed that HPP at low temperature will retain the flavor of the product; however, Hofmann et al. (24) reported that HPP could change the formation of Maillard-derived compounds at high temperature. Information on flavor generation under high pressure is still very limited. The objective of this study was to investigate volatile generation in milk under high pressure and moderate temperature and to compare the volatile formation with that formed under atmospheric pressure conditions at comparable temperature.

Table 1. Experimental Design for Pressure, Temperature, and Time Treatments of Milk Samples treatment parameters

MATERIALS AND METHODS Chemicals. 3-Methylbutanal, 2-methylpropanal, ethyl acetate, 3-methylbutanol, 2-furaldehyde, heptanal, octanal, nonanal, decanal, trans-2-hexenal, 2-heptanone, 2-nonanone, 2-undecanone, 3-heptanone, 3-octanone, 4-decanone, methanethiol (MeSH), dimethyl disulfide (DMDS), dimethyl trisulfide (DMTS), dimethyl sulfoxide (DMSO), dimethyl sulfone (Me2SO2), and isopropyl disulfide (IPDS) were purchased from Aldrich Chemical Co. Inc. (Milwaukee, WI); 2,3butanedione (diacetyl) and hexanal were purchased from Sigma (St. Louis, MO); 2-octanone was from Fluka Chemical Corp. (Milwaukee, WI); 2-pentanone, 2-hexanone, 2-decanone, and trans-2-nonenal were from K&K Laboratories (Jamaica, NY); 2-methylbutanal was from Polyscience Inc. (Niles, IL); dimethyl sulfide (DMS) and ethyl methyl sulfide (EtMeS) were from TCI America (Portland, OR). Carbon disulfide (CS2) was from EMD Chemicals Inc. (Gibbstown, NJ). Milk Samples. Raw homogenized milk with 3.2% fat was obtained locally (Lochmead Farms, Junction City, OR), and sodium azide (0.02%) was added immediately to inhibit microbial activity. The same batch of milk was used for the entire experiment to avoid differences in sample composition. Pasteurized milk samples with 3.2% fat content from two different commercial brands (PA and PB) were purchased locally, stored at 4 °C, and analyzed before their expiration date (2 weeks from manufacturing date). UHT-treated milk samples (3.2% fat) from two different commercial brands (UHT UA and UHT UB) were purchased in Mexico, stored at 4 °C after arrival in the United States, and analyzed before their expiration date (6 months from manufacturing date). High-Pressure Treatments. Raw milk samples were placed in individually sealed polyethylene bags. Treatments from a 2 × 3 × 3 experimental design (treatments 10-27 in Table 1) for temperature (25 and 60 °C), pressure (482, 586, and 620 MPa), and time (1, 3, and 5 min) were run in triplicate. A 2.2 L high-pressure vessel (Engineered Pressure Systems Inc., Haverhill, MA) equipped with a temperature controller and a high-pressure fluid pump (model P100-10FC, HydroPac Inc., Fairview, PA) was used to process the milk. All samples were equilibrated at 25 °C before loading into the high-pressure vessel. Loading time (1 min) and unloading time (1.5 min) were kept constant for all runs. The average pressure-ramp time was 40 s. Time between runs was 15 min. Immediately after treatment, samples were placed in an ice bath and then stored at -38 °C until analyzed. All analyses were performed within 20 days after the treatments. Heat Treatments. An experimental design (treatments 1-9 in Table 1) was run in triplicate under atmospheric pressure to simulate the temperature conditions the sample would experience during a highpressure treatment at 620 MPa at temperatures of 25, 60, and 80 °C and holding times of 1, 3, and 5 min. The treatment conducted at 80 °C was included in the design as an extra level of reference, although it was not used in the high-pressure treatment. These treatments took into account the temperature increase inside the pressure vessel generated during pressurization due to sample compression. To simulate the thermal component of the high-pressure process, samples contained in sealed polyethylene bags were preheated in a water bath at 25, 60, and 80 °C, respectively, for 3.2 min, which represents the total time required in the high-pressure treatments to load the sample (1 min), increase the pressure (0.7 min), and unload the sample (1.5 min) at the preset initial temperature. Immediately after that, the samples were transferred to another water bath and kept at 44, 79, and 99 °C, respectively, for 1, 3, and 5 min, which represents the additional

9185

c

treatment

temperaturea (°C)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 past. Ab past. Bb UHT Ac UHT Bc

25 (43.6) 25 (43.6) 25 (43.6) 60 (78.6) 60 (78.6) 60 (78.6) 80 (98.6) 80 (98.6) 80 (98.6) 25 25 25 25 25 25 25 25 25 60 60 60 60 60 60 60 60 60

pressure (MPa)

time (min)

482 482 482 586 586 586 620 620 620 482 482 482 586 586 586 620 620 620

1 3 5 1 3 5 1 3 5 1 3 5 1 3 5 1 3 5 1 3 5 1 3 5 1 3 5

a Refer to text for treatment detail. b Pasteurized commercial milk samples. Ultrahigh-temperature-processed commercial milk samples.

temperature increase that milk samples would experience under high pressure due to compression heating estimated on an average increase of 3 °C per 100 MPa (25). Immediately after each treatment, samples were placed in an ice bath and then stored at -38 °C until analyzed. Analysis of Thermally Derived Volatile Compounds. Dimethyl sulfide and thermally derived volatile compounds were analyzed using headspace solid-phase microextraction and gas chromatography with flame ionization detection (HS-GC/FID) (8). The milk sample (20 g) was extracted with a 2-cm 50/30 µm divinylbenzene/carboxen/ polydimethylsiloxane (DVB/CAR/PDMS) fiber (Supelco Co., Bellefonte, PA) at 35 °C for 1 h. Volatile compounds were analyzed using an HP 5890 series II gas chromatograph (Hewlett-Packard, Wilmington, DE) equipped with an FID and an HP-5 capillary column (50 m × 0.32 mm i.d., 0.52 µm film thickness, Hewlett-Packard). trans-2Hexenal, 3-heptanone, 3-octanone, 4-decanone, and trans-2-nonenal were used as internal standards. Calibration curves for the volatile compounds were constructed on the basis of the standard addition technique. Analysis of Trace Volatile Sulfur Compounds. Hydrogen sulfide, methanethiol, dimethyl disulfide, dimethyl trisulfide, carbon disulfide, dimethyl sulfoxide, and dimethyl sulfone were analyzed using headspace solid-phase microextraction and gas chromatography with pulsed-flame photometric detection (HS-SPME/GC-PFPD) described previously (9). The milk sample (10 g) was extracted with a 85 µm carboxen/ polydimethylsiloxane (CAR/PDMS) fiber (Supelco) at 30 °C for 15 min and analyzed with a Varian CP-3800 gas chromatograph (Varian Inc., Walnut Creek, CA) equipped with a DB-FFAP fused-silica capillary column (30 m × 0.32 mm, 1.0 µm film; Agilent Technologies, Inc., Palo Alto, CA) and a PFPD detector. Calibration curves for seven sulfur-containing compounds were constructed in milk by the standard addition technique. Ethyl methyl sulfide and isopropyl disulfide were used as internal standards. Triplicate analysis was performed on each sample.

9186

J. Agric. Food Chem., Vol. 54, No. 24, 2006

Statistical Analysis. The concentration of each of the 27 volatile compounds was compared to that of the raw milk sample (3.2% fat), and a normalized percentage change was calculated as follows:

% change )

concn in treated milk - concn in raw milk × 100 concn in raw milk

Multivariate analysis of variance (MANOVA) and principal component analysis (PCA) were conducted using S-Plus 6.2 (Insightful Corp., Seattle, WA) for the normalized percentage of change of 27 compounds as the multivariate response. RESULTS AND DISCUSSION

Effect of Thermal Processing on Volatile Generation in Milk. During high-pressure processing, there is a temperature increase upon pressurization due to adiabatic heating (26). The magnitude of this change depends on the compressibility of the substance and its specific heat. At the pressures typically encountered during high-pressure processing (400-1000 MPa), milk increases ∼3 °C for every 100 MPa of pressure increase (25). The temperature decreases instantly when the pressure is released. Because thermal degradations of lipids, proteins, and sugars contribute most to the volatile formation in milk, the adiabatic heating effect was taken into account for all samples treated under atmospheric pressure, so the corresponding treatments submitted to high pressure could be directly compared. Treatments 1-9 (Table 1) simulated the temperature increase experienced inside the pressure vessel at 620 MPa. Twenty-seven volatile compounds including aldehydes, ketones, esters, and sulfur compounds were quantified in this study. Because the concentrations of volatile compounds varied widely and were affected by treatment conditions, the normalized percentage change for each compound was also investigated. As shown in Table 2, the concentrations of aldehydes and ketones were at the microgram per kilogram level, with hexanal, nonanal, and decanal dominating for most of the treatments. The concentration of both straight-chain and branched-chain aldehydes increased with the severity of the applied heat treatment (Table 2). At 25 °C, the concentrations of aldehydes in all treatments were the same as those in the raw milk, demonstrating that none of the aldehydes was generated at this temperature. When the temperature was raised to 60 °C, the concentrations of nonanal in the heated samples were triple those in the raw milk, whereas the concentrations of octanal and decanal were double. Other aldehydes also increased but at different degrees. Hexanal, on the other hand, had lower concentrations than that in the raw milk. More dramatic increases of aldehydes were observed at 80 °C. In all cases, nonanal had the highest relative increase upon heating, followed by octanal and decanal. At both 60 and 80 °C, longer holding time increased the formation of aldehydes. Straight-chain aldehydes can result from the autoxidation of unsaturated fatty acids and spontaneous decomposition of hydroperoxides promoted by heat (27). At low temperature, the increase of C2-7,9 aldehydes is thought to be the main cause for the stale offflavor in milk (8, 28, 29). At high temperature, they probably also contribute to the “cooked” off-flavor because the concentrations of most aldehydes were higher than their sensory thresholds (29). Branched-chain aldehydes are from Strecker degradation of amino acids. Methyl ketones were also dramatically affected by thermal treatment (Table 2A), which is in agreement with a previous study (7). Similar to aldehydes, methyl ketones were not formed

Vazquez-Landaverde et al. at 25 °C. When samples were heated at 60 °C, the concentrations of 2-nonanone increased approximately 4 times, whereas 2-undecanone and 2-hexanone increased 2 times. At 80 °C, 2-decanone and 2-heptanone also had dramatic increases. The concentrations of 2-octanone were not affected by thermal treatments. Although methyl ketones are naturally present in raw milk, most of them are formed thermally by the oxidation of fatty acids (30) or by the decarboxylation of β-ketoacids naturally present in milk fat (27, 31). 2,3-Butanedione and 2-furaldehyde also showed an increasing trend in concentration with temperature and heating time (Table 2B). Both of them can be formed through the thermal degradation of carbohydrates. Ethyl acetate had slight increases only at 80 °C, and it is formed by esterification of ethanol with acetic acid catalyzed by heat (32). 3-Methylbutanol was also monitored; this compound is naturally present in raw milk and is produced mainly by the microbial reduction of 3-methylbutanal (33). However, all of these compounds, with the exception of 2,3-butanedione, have very high sensory thresholds (29), suggesting that they are of little importance for the development of off-flavors in heated milk. The concentrations of sulfur-containing compounds presented the widest variation. Dimethyl sulfoxide and dimethyl sulfone were the most abundant, reaching milligram per kilogram levels, whereas CS2, DMDS, and DMTS were present only at nanogram per kilogram levels. It has been reported that volatile sulfur compounds are mainly responsible for the development of the “cooked” off-flavor in heated milk (2, 34-36). A recent study based on odor activity value (ratio of concentration to sensory threshold) further suggested that MeSH, DMS, DMTS, and H2S are probably the most powerful off-flavor compounds in heated milk (8, 9). Most volatile sulfur compounds followed an increasing trend in concentration when temperature and processing time were increased (Table 2B). H2S and MeSH had dramatic increases upon heating. At 60 °C, the concentration of MeSH increased 3-4 times depending on the holding time, whereas H2S increased 3-7 times. This increase was even more dramatic at 80 °C; an increase of 8-9-fold was observed for MeSH. An increase of 10-15-fold was noted for H2S, raising its concentration comparable to that of MeSH. Although H2S had the highest concentration increase upon heating, its contribution to the offflavor of heated milk is probably not as important as previously suggested (28, 36, 37). H2S has a sensory threshold of 10 µg/ kg in water (29); on the basis of its calculated odor activity value, its contribution to “heated” off-flavor is probably limited to samples subjected to only the most severe temperature-time treatments (60 °C for 5 min or 80 °C). Even with the most severe temperature-time treatment, it had odor activity values of only 2-3. The concentration of MeSH in all treatments far exceeded its sensory threshold (0.2 µg/kg in water) (29), and its odor activity value increased from 15 in raw milk to as high as 140 in heated milk (80 °C/5 min), which further confirms that MeSH may be a much more important off-flavor contributor than H2S (9). DMS had a slight increase in concentration upon heating, and its concentration in all treatments was higher than its sensory threshold (2 µg/kg in water) (29). Calculated odor activity values (OAV 2-4) suggest that it may contribute to the off-flavor of heated milk but to a much lesser degree than MeSH. Although increasing upon heating, concentrations of DMTS in all treatments were below its sensory threshold; thus, its contribution to the off-flavor of heated milk is probably very limited, if any. The concentrations of DMDS were also below its sensory threshold.

J. Agric. Food Chem., Vol. 54, No. 24, 2006

Volatile Profile of Milk

9187

Table 2. Concentrations (Micrograms per Kilogram) of (A) Straight-Chain Aldehydes and Methyl Ketones in Milk and (B) Volatile Sulfur and Other Compounds in Milka (A) Straight-Chain Aldehydes and Methyl Ketones treatmentb

2-pentanone

2-hexanone

hexanal

2-heptanone

heptanal

2-octanone

octanal

2-nonanone

nonanal

2-decanone

decanal

2-undecanone

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 raw past. A past. B UHT A UHT B

0.73 0.87 0.66 0.84 0.96 1.99 1.85 1.69 1.77 0.43 0.54 0.48 0.59 0.55 0.41 0.54 0.49 0.43 0.94 0.91 1.01 0.83 0.87 0.83 0.84 0.74 0.92 0.75 0.21 0.22 9.51 9.68

0.66 1.05 1.08 1.74 1.56 1.68 1.74 1.72 1.63 0.40 0.62 0.50 0.54 0.48 0.45 0.51 0.51 0.42 1.21 1.41 1.09 1.61 1.19 1.08 1.62 1.46 1.64 0.93 0.05 0.15 1.82 1.93

14.40 12.42 13.48 10.33 11.48 11.87 13.23 15.24 16.98 9.74 10.63 8.09 8.64 6.46 6.19 8.70 9.47 7.88 11.85 14.94 12.04 15.08 19.25 23.88 13.91 15.71 38.16 13.43 8.83 7.72 12.76 13.52

1.04 1.04 1.08 1.12 1.22 1.57 1.94 3.23 5.24 1.08 1.01 1.06 1.35 1.02 1.01 1.09 1.04 1.24 2.71 2.05 1.92 2.03 1.84 2.36 2.07 1.91 1.98 1.05 1.11 0.76 32.46 35.62

1.13 0.81 1.21 1.55 1.53 1.78 1.86 2.14 2.61 1.02 1.09 1.05 1.01 0.92 0.75 1.01 1.06 0.86 1.17 1.59 1.40 1.76 3.12 3.12 1.60 2.22 5.63 1.05 0.06 0.14 1.68 1.73

4.31 3.91 4.09 4.06 3.84 3.90 4.35 4.79 4.82 4.21 5.00 3.87 5.07 4.08 2.96 4.34 4.89 3.71 3.15 4.44 4.35 3.70 4.14 4.01 4.79 3.74 4.39 4.11 6.35 3.02 4.53 4.65

1.49 1.33 1.49 2.31 2.35 2.73 3.29 3.74 5.13 1.00 1.18 0.92 1.06 1.00 0.74 1.03 1.00 0.88 1.47 1.57 1.53 1.57 1.83 1.92 1.43 1.30 2.35 1.44 0.13 0.14 0.91 1.02

0.77 0.76 1.10 3.03 3.22 3.80 4.61 6.29 8.64 0.50 0.58 0.54 0.36 0.63 0.52 0.60 0.59 0.45 2.07 2.07 2.31 2.63 2.37 2.05 1.36 1.40 2.03 0.88 0.51 0.62 51.72 54.28

3.40 3.42 4.59 9.01 12.74 13.67 17.09 20.75 26.56 2.70 2.97 3.46 2.32 3.01 2.57 2.85 3.72 2.34 5.59 5.95 5.96 6.31 7.43 7.54 8.54 6.46 11.30 3.80 1.18 0.42 3.91 4.06

0.51 0.50 0.57 0.67 0.80 0.76 1.05 1.05 1.12 0.40 0.41 0.51 0.38 0.64 0.42 0.58 0.80 0.34 0.91 1.10 1.06 0.99 1.06 0.93 0.71 0.63 0.80 0.53 0.21 0.20 1.35 1.38

9.68 5.99 8.47 16.11 16.83 19.01 21.60 21.79 23.69 4.28 4.52 5.20 5.28 5.89 3.83 7.32 8.82 4.87 15.65 16.64 14.36 12.27 16.47 14.80 10.60 11.46 14.52 8.05 5.91 1.67 6.56 6.85

0.31 0.36 0.36 0.33 0.64 0.63 1.09 1.37 1.57 0.28 0.46 0.43 0.33 0.32 0.21 0.39 0.49 0.27 0.51 0.47 0.55 0.47 0.47 0.61 0.48 0.37 0.53 0.34 2.73 0.91 9.63 10.10

treatmentb

2-methylpropanal

2,3-butanedione

ethyl acetate

3-methylbutanal

2-methylbutanal

3-methyl1-butanol

2-furaldehyde

H2S

MeSH

DMS

DMDS (ng/kg)

DMTS (ng/kg)

CS2 (ng/kg)

DMSO

Me2SO2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 raw past. A past. B UHT A UHT B

0.33 0.38 0.38 0.35 0.42 0.48 0.59 0.73 0.76 0.31 0.45 0.50 0.54 0.35 0.23 0.40 0.43 0.40 0.44 0.51 0.44 0.58 0.45 0.46 0.47 0.49 0.47 0.36 0.51 0.28 2.51 2.63

1.34 1.44 1.50 1.78 1.65 1.79 1.95 2.39 2.82 1.28 1.32 1.43 1.41 0.94 0.69 0.82 1.19 1.19 2.15 2.35 2.11 2.24 1.65 1.87 2.05 1.84 1.98 1.42 0.93 2.09 7.37 7.52

0.28 0.22 0.23 0.25 0.30 0.31 0.40 0.38 0.39 0.26 0.33 0.32 0.31 0.34 0.18 0.32 0.32 0.27 0.39 0.39 0.30 0.26 0.27 0.32 0.47 0.41 0.32 0.25 0.33 0.43 2.17 2.18

1.14 1.70 1.69 1.96 1.83 2.77 2.87 1.29 2.08 0.53 0.83 0.99 0.96 0.99 0.50 0.89 0.85 0.78 1.86 2.24 1.88 2.12 1.93 1.81 2.21 1.83 2.09 1.51 0.05 0.09 1.14 1.26

5.06 6.63 6.37 7.34 5.50 9.12 8.21 8.07 7.66 2.42 3.43 3.38 3.55 2.62 2.14 3.12 3.17 3.01 5.13 6.32 6.31 6.91 6.47 5.11 6.16 5.26 6.68 6.02 0.08 0.12 0.91 1.12

0.39 0.51 0.46 0.55 0.45 0.67 0.71 0.66 0.68 0.24 0.30 0.33 0.31 0.31 0.21 0.31 0.31 0.27 0.46 0.55 0.55 0.58 0.54 0.50 0.59 0.53 0.54 0.45 0.13 0.13 0.14 0.16

0.98 1.99 2.07 3.20 1.34 3.28 3.77 2.99 3.34 0.14 0.45 0.62 0.76 0.73 0.71 0.71 0.59 0.64 1.74 2.67 1.80 2.40 1.82 1.60 2.27 1.99 2.78 1.68 0.12 0.11 0.37 0.41

1.83 2.01 1.95 1.70 7.15 16.43 21.46 25.35 30.25 5.78 8.09 6.19 1.71 3.18 7.33 6.66 6.48 11.94 12.37 13.69 8.01 6.81 9.02 9.53 14.47 15.68 17.95 1.93 0.86 1.02 12.25 12.23

3.41 2.77 3.24 4.48 9.18 12.66 22.37 24.87 28.36 3.40 7.60 5.78 5.02 6.16 3.41 2.74 3.63 3.20 7.08 6.52 5.68 7.90 8.50 8.76 4.25 4.44 3.26 3.14 3.59 4.28 23.92 24.65

5.27 5.60 5.51 5.33 5.64 5.73 7.99 8.44 8.70 5.23 7.07 5.99 6.61 6.58 5.96 6.91 6.52 5.99 6.13 6.46 7.65 6.09 6.36 6.30 5.93 5.93 6.83 5.46 8.47 16.08 21.12 22.42

61 80 76 63 66 38 89 60 44 15 20 34 95 55 19 29 41 20 27 31 23 33 31 30 14 19 32 72 33 17 33 33

42 36 42 39 46 48 49 65 72 24 30 35 36 32 19 34 35 29 27 36 38 34 33 31 34 38 41 40 17 19 47 49

36 33 33 36 42 50 44 43 65 135 108 113 104 105 129 147 104 120 49 65 39 54 47 63 64 60 69 34 30 33 69 60

620 638 606 647 702 748 872 1074 1241 398 592 867 757 676 520 678 759 765 967 735 559 915 831 992 711 731 683 621 696 790 1427 1478

1653 1480 1412 1328 1481 1776 1269 1273 2143 1166 1199 1725 1679 1171 1051 1528 1798 1719 1734 1518 1596 1507 1414 1920 1704 1552 1367 1515 1346 2224 1268 1275

(B) Volatile Sulfur and Other Compounds

a

Each value represents the average of a triplicate treatment. Relative standard deviation was 0 has the opposite effect. The sensitivity of a chemical reaction to pressure will increase with the absolute value of ∆V* (43). The determination of ∆V* values is not possible with the experimental data available in this study and will require more comprehensive kinetic studies. The formation of H2S seemed to be affected by both pressure and holding time. A dramatic increase of H2S was observed under high-pressure treatments even at 25 °C (Table 2B). The increases seemed to be holding time dependent. As expected, H2S increased even more at 60 °C, and its concentrations in high-pressure-treated samples were generally higher than in the corresponding heat-only samples. Although the formation of H2S could be pressure dependent, it was not obvious, however, because the trend was not consistent for all treatments. The concentrations of MeSH also increased at 25 °C under high pressure. However, when the pressure increased to 620 MPa, the concentration of MeSH decreased. The same behavior was observed at 60 °C, with the lowest concentration occurring at 620 MPa and 5 min of holding. Although methanethiol formation appeared to be inhibited under pressure, it is also possible that it was converted to other compounds. In addition, the formation and conversion of methanethiol could be pH dependent, and the effect of pH on volatile formation was not taken into consideration in this study, because pressure-induced pH shifts cannot be measured in the experimental HPP vessel as there are no pressure-resistant probes currently available. There was only a slight increase for DMS under high pressure, and this increase was undistinguishable from 25 to 60 °C. The concentrations of other sulfur compounds were similar to that

Figure 2. PCA plot for the volatile profile of milk subjected to different high hydrostatic pressure and temperature treatments (treatments 10− 27).

of the homologous heat treatment. Many sulfur compounds are extremely reactive, and more studies are needed to understand their formation behavior under high pressure. The impact of HPP on volatile formation can be better illustrated by MANOVA on normalized percentage change. From the results of MANOVA, it can be concluded that

9190

J. Agric. Food Chem., Vol. 54, No. 24, 2006

Figure 3. Spatial distribution of volatile profile of milk subjected to temperature and high hydrostatic pressure−temperature treatments.

temperature, pressure, and time, as well as their double and triple interactions, were important factors influencing the formation of volatile compounds in milk (P < 0.001). The PCA plot for the pressurized samples (Figure 2) confirmed that temperature had an important effect, separating the samples treated at 25 °C (group 1) from those treated at 60 °C (groups 2 and 3). At 25 °C, neither pressure nor holding time appeared to affect the volatile profile of milk, as all treatments at 25 °C were grouped closely together (group 1). However, when temperature was increased to 60 °C, pressure became an important factor separating samples treated at 620 MPa (group 3) from those treated at 586 and 482 MPa (group 2) at the same temperature. The results demonstrated that pressure had an impact on the volatile generation of milk but only at moderately high temperature (60 °C). Samples treated at 60 °C and 620 MPa were the only ones clearly scattered due to holding time (group 3), therefore suggesting that time has an important effect only at high levels of pressure and temperature. To understand how the pressure and temperature will affect the volatile formation in milk, the spatial distribution of all HPPtreated and non-HPP-treated milk samples was plotted in Figure 3. Milk treated under milder conditions, either with pressure below 620 MPa or with temperature below 60 °C, formed one cluster (group 1), indicating that the milk volatile profile did not change significantly under these conditions. Heat-treated samples at 60 °C (group 2) and 80 °C (group 4) formed clusters clearly separated from group 1. On the other hand, those treated at 60 °C and 620 MPa formed a different cluster (group 3). It was interesting to note that the samples in group 3 were clustered in the opposite side of their heat-treated homologues of group 2 in the PCA plot. The PCA loadings for the compounds (Table 3) showed that the major changes in group 3 (620 MPa, 60 °C) were mostly defined by an increase in concentrations of H2S, hexanal, heptanal, and nonanal, which were different from their corresponding heat-only treatments (group 2), where the greatest increases were for H2S, MeSH, 2-nonanone, nonanal, 2-undecanone, and 2-heptanone. Figure 4 compares the volatile profiles of milk samples under all pressure, temperature, and time treatments with two com-

Vazquez-Landaverde et al.

Figure 4. Comparison of volatile profile in commercially processed milk with samples subjected to temperature and high-pressure treatments.

mercial pasteurized and two commercial UHT samples. Although these commercially processed samples were not obtained from the same source of raw milk used for the experiments, they could provide a relative comparison between the highpressure treatments and commercial thermal processing. Pasteurized samples were located inside a major cluster (Figure 4, group 1) that included the milder treatments of heating (60 °C and lower) and all pressurization runs except those at the highest pressure and temperature combinations (620 MPa and 60 °C). From this PCA plot (Figure 4), in addition to previous studies on off-flavor compounds in heated milk (8, 9), it could be inferred that all of the samples in group 1 have aroma profiles similar to that of pasteurized milk. Samples located in groups 2 and 3 were those submitted to the highest levels of heat and pressure treatments and could present aroma profiles different from those of samples in group 1. UHT milk samples (group 4) were located far from any other cluster in the PCA chart, because they had the highest concentration of volatile compounds and could impart a “cooked” off-flavor note, yet further sensory analysis is needed to prove this prediction. In summary, high-pressure processing at low temperature causes minimum change of the volatile composition of milk. However, under extreme pressure and temperature conditions, volatile compound formation is different from that under atmospheric pressure conditions. Heat treatment at high temperature promotes the formation of both aldehydes and methyl ketones, whereas high pressure at high temperature favors the formation of aldehydes. The formation of sulfur compounds was also different under high pressure. Further determinations of the kinetic behavior of these compounds may help us to understand their formation under high hydrostatic pressure. ACKNOWLEDGMENT

We thank Clifford Pereira from the Statistics Department of Oregon State University for valuable advice in the statistical analyses performed in this study.

Volatile Profile of Milk LITERATURE CITED (1) Shipe, W. F. Analysis and control of milk flavor. In The Analysis and Control of Less Desirable FlaVors in Foods and BeVerages; Charalambous, G., Ed.; Academic Press: New York, 1980; 201 pp. (2) Steely, J. S. Chemiluminiscence detection of sulfur compounds in cooked milk. In Sulfur Compounds in Foods; Mussinan, C. J., Keelan, M. E., Eds.; American Chemical Society: Washington, DC, 1994; pp 22-35. (3) Scanlan, R. A.; Lindsay, R.; Libbey, L. M.; Day, E. A. Heatinduced volatile compounds in milk. J. Dairy Sci. 1968, 51, 1001-1007. (4) Jeon, I. J.; Thomas, E. L.; Reineccius, G. A. Production of volatile flavor compounds in ultrahigh-temperature processed milk during aseptic storage. J. Agric. Food Chem. 1978, 26, 1183-1188. (5) Moio, L.; Etievant, P. X.; Langlois, D.; Dekimpe, J.; Addeo, F. Detection of powerful odorants in heated milk by use of extract dilution sniffing analysis. J. Dairy Res. 1994, 61, 385394. (6) Contarini, G.; Povolo, M.; Leardi, R.; Toppino, P. M. Influence of heat treatment on the volatile compounds of milk. J. Agric. Food Chem. 1997, 45, 3171-3177. (7) Contarini, G.; Povolo, M. Volatile fraction of milk: comparison between purge and trap and solid phase microextraction techniques. J. Agric. Food Chem. 2002, 50, 7350-7355. (8) Vazquez-Landaverde, P. A.; Velazquez, G.; Torres, J. A.; Qian, M. C. Quantitative determination of thermally derived volatile compounds in milk using solid-phase microextraction and gas chromatography. J. Dairy Sci. 2005, 88, 3764-3772. (9) Vazquez-Landaverde, P. A.; Torres, J. A.; Qian, M. C. Quantification of trace volatile sulfur compounds in milk by solidphase microextraction and gas chromatography-pulsed flame photometric detection. J. Dairy Sci. 2006, 89, 2919-2927. (10) Fang, Y.; Qian, M. C. Sensitive quantification of sulfur compounds in wine by headspace solid-phase microextraction technique. J. Chromatogr. A 2005, 1080, 177-185. (11) Burbank, H. M.; Qian, M. C. Volatile sulfur compounds in Cheddar cheese determined by headspace solid-phase microextraction and gas chromatograph-pulsed flame photometric detection. J. Chromatogr. A 2005, 1066, 149-157. (12) Elwell, M. W.; Barbano, D. M. Use of microfiltration to improve fluid milk quality. J. Dairy Sci. 2006, 89, 20-30. (13) Rysstar, G.; Kolstad, J. Extended shelf-life milk: advances in technology. Int. J. Dairy Technol. 2006, 59, 85-96. (14) Torres, J. A.; Velazquez, G. Commercial opportunities and research challenges in the high pressure processing of foods. J. Food Eng. 2004, 67, 96-112. (15) Berlin, D. L.; Herson, D. S.; Hicks, D. T.; Hoover, D. G. Response of pathogenic Vibrio species to high hydrostatic pressure. Appl. EnViron. Microbiol. 1999, 65, 2776-2780. (16) Cheftel, J. C. High pressure, microbial inactivation and food preservation. C. R. Acad. Agric. Fr. 1995, 81, 13-38. (17) Velazquez, G.; Gandhi, K.; Torres, J. A. High hydrostatic pressure: a review. Biotam 2002, 12, 71-78. (18) Serrano, J.; Velazquez, G.; Lopetcharat, K.; Ramirez, J. A.; Torres, J. A. Effect of moderate pressure treatments on microstructure, texture, and sensory properties of stirred-curd Cheddar shreds. J. Dairy Sci. 2004, 87, 3172-3182. (19) Serrano, J.; Velazquez, G.; Lopetcharat, K.; Ramirez, J. A.; Torres, J. A. Moderately hydrostatic pressure processing to reduce production costs of shredded cheese: microstructure, texture and sensory properties of shredded milled curd Cheddar. J. Food Sci. 2005, 70, 286-293. (20) Rademacher, B.; Kessler, H. G. Presented at the European High Pressure Research Conference, Leuven, Belgium.

J. Agric. Food Chem., Vol. 54, No. 24, 2006

9191

(21) Tovar-Hernandez, G.; Pen˜a, H. R. V. G.; Ramirez, J. A.; Torres, J. A. Presented at the IFT Annual Meeting, New Orleans, LA. (22) Hinrichs, J.; Kessler, H. G. Presented at the European High Pressure Research Conference, Reading, U.K. (23) Buchheim, W.; El-Nour, A. M. A. Induction of milkfat crystallization on the emulsified state by high hydrostatic pressure. Fat Sci. Technol. 1996, 94, 369-373. (24) Hofmann, T.; Deters, F.; Heberle, I.; Schieberle, P. Influence of high hydrostatic pressure on the formation of Maillard-derived key odorants and chromophores. Ann. N. Y. Acad. Sci. 2005, 1043, 893. (25) Ting, E.; Balasubramaniam, V. M.; Raghubeer, E. Determining thermal effects in high-pressure processing. Food Technol. 2002, 56, 31-56. (26) Harvey, A. E.; Peskin, A. P.; Klein, S. A. NIST/ASME Steam Program; Physical and Chemical Properties Division, National Institute of Standards and Technology, U.S. Department of Commerce: Boulder, CO, 1996. (27) Grosch, W. Lipid degradation products and flavour. In Food FlaVours. Part A. Introduction; Morton, I. D., Macleod, A. J., Eds.; Elsevier Scientific Publishing: Oxford, U.K., 1982; pp 325-385. (28) Rerkrai, S.; Jeon, I. J.; Bassette, R. Effect of various direct ultrahigh temperature heat treatments on flavor of commercially prepared milks. J. Dairy Sci. 1987, 70, 2046-2054. (29) Rychlik, M.; Schieberle, P.; Grosch, W. Compilation of Odor Thresholds, Odor Qualities and Retention Indices of Key Food Odorants; Deutsche Forschungsanstalt fu¨r Lebensmittelchemie and Institut fu¨r Lebensmittelchemie der Technischen Universita¨t Mu¨nchen: Garching, Germany, 1998. (30) Nawar, W. W. Lipids. In Food Chemistry; Fennema, O. R., Ed.; Dekker: New York, 1996; pp 290-291. (31) Jensen, R. G.; Bitman, J.; Carlson, S. E.; Couch, S. C.; Hamosh, M.; Newburg, D. S. Milk lipids. In Handbook of Milk Composition; Jensen, R. G., Ed.; Academic Press: San Diego, CA, 1995; pp 571-572. (32) Hart, H. Organic Chemistry: A Short Course, 8th ed.; Houghton Mifflin: Boston, MA, 1991; pp 283-284. (33) Toso, B.; Procida, G.; Stefanon, B. Determination of volatile compounds in cows’ milk using headspace GC-MS. J. Dairy Res. 2002, 69, 569-577. (34) Simon, M.; Hansen, A. P. Effect of various dairy packaging materials on the shelf life and flavor of ultrapasteurized milk. J. Dairy Sci. 2001, 84, 784-791. (35) Datta, N.; Elliot, A. J.; Perkins, M. L.; Deeth, H. C. Ultra-high temperature (UHT) treatment of milk: comparison of direct and indirect modes of heating. Aust. J. Dairy Technol. 2002, 57, 211-227. (36) Christensen, K. R.; Reineccius, G. A. Gas chromatographic analysis of volatile sulfur compounds from heated milk using static headspace sampling. J. Dairy Sci. 1992, 75, 2098-2104. (37) Badings, H. T.; Van der Pol, J. J. G.; Neeter, R. Aroma compounds which contribute to the difference in flavor between pasteurized milk and UHT milk. In FlaVor; Schreider, P., Ed.; de Gruyter: Berlin, Germnay, 1981; pp 683-692. (38) Shibamoto, T.; Mihara, S.; Nishimura, O.; Kamiya, Y.; Aitoku, A.; Hayashi, J. Flavor volatiles formed by heated milk. In The Analysis and Control of Less Desirable FlaVors in Foods and BeVerages; Charalambous, G., Ed.; Academic Press: New York, 1980; pp 260-263. (39) Damodaran, S. Amino acids, peptides, and proteins. In Food Chemistry; Fennema, O. R., Ed.; Dekker: New York, 1996; pp 412-413. (40) Zheng, Y.; Ho, C. Kinetics of the release of hydrogen sulfide from cysteine and glutathione during thermal treatment. In

9192

J. Agric. Food Chem., Vol. 54, No. 24, 2006

Sulfur Compounds in Foods; Mussinan, C. J., Keelan, M. E., Eds.; American Chemical Society: Washington, DC, 1994; pp 138-146. (41) Galazka, V. B.; Ledward, D. A. Effects of high pressure on protein polysaccharide interactions. In Macromolecular Interactions in Food Technology; Parris, N., Kato, A., Creamer, L. K., Pearce, J., Eds.; American Chemical Society: Washington, DC, 1996; pp 113-123. (42) McNaught, A. D.; Wilkinson, A. IUPAC Compendium of Chemical Terminology, 2nd ed.; Blackwell Science: Cambridge, U.K., 1997.

Vazquez-Landaverde et al. (43) Mussa, D.; Ramaswamy, H. Ultra high pressure pasteurization of milk: kinetics of microbial destruction and changes in physico-chemical characteristics. Lebensm. Wiss. Technol. 1997, 30, 551-557. Received for review May 27, 2006. Revised manuscript received September 12, 2006. Accepted September 17, 2006. This project was financially supported by Dairy Management Inc. (Rosemont, IL), and P.A.V.-L. received Scholarship 142410 from CONACYT (Mexico).

JF061497K